BACKGROUND PROBLEM/RATIONALE: Tumor growth in athymic nude mice is one of the most important experimental tools used to study the biology and treatment of human cancer. Until recently, it was felt this methodology suffered from the failure of such xenografts to metastasize, or to do so in a clinically relevant manner. This problem has been partly overcome, for some types of tumors, with the implementation of orthotopic transplantation techniques. However, this is not the case for malignant melanomas, which rarely manifest extra-pulmonary metastases in a tissue-specific manner. A possible solution to this problem was proposed through the isolation of lectin-resistant mutants of human melanoma cells. This is based on the hypothesis that stable `tissue-specific' metastatic melanoma subpopulations exist in very low frequencies, but can be isolated and enriched through selection of certain types of Lecr mutants, i.e. cells which possess specific alterations in cell surface glycoconjugates. RECENT FINDINGS/EXPERIMENTAL APPROACH: Using a long-term multi- step selection protocol several remarkable spontaneous wheat germ agglutinin-resistant (WGRr) variant sublines of a human melanoma cell line (MeWo) were recently isolated. One of these (70-W) manifested multiple bilateral cerebral, skin, abdominal, gut and lung metastases after iv inoculation of the cells. Moreover, the metastases were always intensely melanotic, in contrast to the MeWo parent - which is mostly amelanotic and metastasizes exclusively to the lungs. A second WGAr subline (3S5) was obtained which was found to be virtually devoid of pulmonary metastatic capacity. Both were found to be phenotypically stable for prolonged periods in culture in the absence of WGA. LONG-TERM OBJECTIVES AND SIGNIFICANCE: The availability of these and new variants presents a unique opportunity to rigorously analyze the genetic, biochemical, and biological basis of human malignant melanoma metastasis and its experimental treatment. Furthermore, their behavior increases the likelihood that the results will be relevant to melanoma in man. The long-term objective of the proposed studies is to determine the basis for the malignant characteristics of the cell lines. Experiments are proposed to genetically characterize the variants using somatic cell and molecular genetic, including gene transfer, technology. Biochemical analysis will be undertaken to determine the nature of glycosylation changes in the cells and their functional significance. Finally, on the basis of several hypotheses, attempts will be made to identify the existence of tissue-specific growth factors, or adhesion molecules which may help determine the metastatic tissue trophism of the 70-W variant.